Statistical quantity based reversible watermarking for copyright protection of digital images

Abstract

Digital watermarking has been proposed as a solution to the problem of resolving copyright ownership of multimedia data (image, audio and video). Devising effective watermarking schemes especially the security oriented applications like copyright protection, copy control, etc., is an extremely challenging task. Most of the existing reversible watermarking techniques have much difficulty in practical applicability due to lack of sufficient reversibility in large scale image datasets, poor invisibility and in some cases robustness is destroyed by unintentional attacks. All this have to be addressed in an efficient technique. The work presented in this paper is concerned with the design of robust digital image watermarking algorithms for copyright protection. In the embedding part, haar wavelet transform is applied to the cover and the watermark image in which the lower subband coefficients of cover image are statistically analyzed for selection of watermark embedding. To embed the watermark bits, the watermark strength is adjusted using enhanced pixel-wise masking. In the extraction part, an efficient inverse detection algorithm is used for extracting the watermark image. It is experimentally confirmed that this technique gives excellent results under tested image processing attacks and JPEG compression.